andrewdoherty

iTrader: (49)

Well most of this technical data has gone over my head, but the basic commutations and accuracy/repeat-ability, temp stability concepts make all the sense in the world. Just for an intellectual exercise, wouldn't accuracy be improved if 6 sensors were used and each pair were averaged for each rotation? Are esc electronics capable of calculating that average in time to make it useful?

Nerobro

iTrader: (2)

Maybe? to what end? What would provide better precision, is using jigs to glue down the components before soldering.

You could do averaging... But if you've got a 32 bit micro on the ESC, you can just "do the math". You know the poles of the stator are 120deg apart, the poles of the rotor are 180deg. The motor could outpulse a standard rpm.. and then easily measure the deviation from the norm and build up a compensation chart.

Adding more sensors, really just puts you in the bucket with "are these in a good place too?" So math and calibration might be the thing in order.

I'd need to look at the ROAR rules more closely, but I suspect using synthetic timing events would get you outside of the Blinky rules.

I'd go back to a Jig I think. That is probally the best option, and it's ESC independant.

howardcano

iTrader: (37)

If running in non-timing ("Blinky") mode, ROAR requires the drive to the motor terminals to change state within 10us of the sensor signal change, all the way up to free-running RPM. Sensorless operation is only allowed in non-blinky classes.

Very few rotors have the poles within 1 degree of 180 apart. Good rotor symmetry is one thing that guys will pay extra to get, as it improves performance. Bad rotor symmetry will throw off the timing calculations of some motor analyzers.

gigaplex

iTrader: (2)

I can't speak about all ESCs, but I have seen some that have an option to switch between full sensored mode, or a hybrid mode where it does switch to back EMF at higher RPM.

Jorge T

iTrader: (14)

For those looking for sensor part numbers.

I see P24ea stamped on a TSR 1.0 sensor board.

A speed passion V3 +30* board has a upside down V symbol in front of G6S.

The 0* SP v3 board has a 1740 stamped on it.

Nerobro

iTrader: (2)

Nice! so they thought of it.

I hadn't considered that rotors might not get magnetized symmetrically. I guess a calibrated ESC might not be a bad idea. It could compensate for the leading or lagging magnetic field of the rotor.

Totally not something i'm gonna do. But there ya go, a way to sell better ESCs.

howardcano

iTrader: (37)

With a little help from yours truly.

I have some that are so bad I don't even need to put them on a motor analyzer. It's very obvious on an oscilloscope, even when the horizontal scale covers a full stator rotation.

Jorge T

iTrader: (14)

All the newer rotors I have are between 0-2* assymetry.( Motiv, Reedy, R1)

All the old geneation rotors I have D3.5/4 are 4-8* and with one tuquois ht at 37* assymetry . I never
could use it. Never new why It made the motor run like cr*p until I checked when I got the motolyser.

Xpress

iTrader: (3)

You just make up for the poor symmetry by running bigger wire....

Nerobro

iTrader: (2)

How does bigger wire compensate for that?

PutAwayWet

iTrader: (1)

I think he is amusing himself with snark regarding the D3.5 debacle. As in: QC doesn't matter if you're willing to simply ignore the rules.

It's amusing.

Now.

Wasn't so much, then.

Nerobro

iTrader: (2)

Hmm. I think I missed that cheating event. Did Trinity just use some illegal wire?

bdmpastx

iTrader: (4)

The Trinity D3.5 Series motor wire was out of spec so ROAR made them illegal and they sold a shit ton of them. We as consumers were screwed. Everyone owned them because they were the motor to have at the time.